Nicely found and shared, Andre!
A couple points to keep in mind when comparing your own figures to published ones:
• Beam brightness varies across the field of illumination, so you may need to take an average rather than peak. Some makers report only peak values.
• Light falls off as the inverse square of the distance from the source so it is best to take your average readings at the same distance as the maker for comparable results.
Normalizing the manufacturer's claims against your meter is really simple. It's my bedtime, so I'll just channel Jeff Liebermann, always a reliable engineer, on this one:
"...if you draw a 1.13meter (44.5inch)
diameter circle on the wall, and shine your bicycle head light or
flashlight on the wall until it fills the circle, the lux meter will
read directly in lumens. It works because the area of the circle is 1
square-meter and:
1 lux = 1 lumen/sq-meter
If your light produces a non-circular spot on the wall or an uneven
pattern, just guess(tm) until you have a spot that looks roughly like
1 square-meter."
The trademarked "guess (tm)" is Jeff's little joke. By the way, it turns out this is a meter that Jeff has discovered on his own and already recommended on a conference known for its cheapskatery and dumpster-diving, so I was right in thinking the price is exceptionally low for such an instrument.
Jeff's trick appears to me to be a) an excellent DIY method of calibrating your meter under the circumstances in which you will use it (the equivalent of touching two DVM probes together to find zero), b) normalizing different lamps against each other, and c) finding your way through the various light-output measuring systems and also d) through the marketing department BS of measuring under only the most favorable conditions and reporting only in an entirely incomparable manner. (That tunnel that BUMM uses is a particularly clever marketing gimmick, because it reflects, indeed multi-bounces, a substantial amount of light.)
This method of normalization has the advantage that you're comparing like with like, and reducing the remaining subjective factors (preferred throw and spread, legality of the lamps, relative price) to fewer than before and more easily resolved -- because you have experience in all these fields -- than total output claims.
If you want to fact-check the manufacturer of your lamp, the square meter test should match his peak output claim and should be higher than any average reading the more honest lamp manufacturers, if any, report.
It is also worth considering that the manufacturer's test, as a matter of practicality, was probably taken with a stiff 6V battery box attached, rather than a hub dynamo. And so were the photographs, with a stationary bike sucking on the battery, not the hub.
